The overall goals of this project are to characterize the determinants of vascular structure and function, and evaluate their effects on cardiovascular morbidity and mortality 1. Increased aortic pulse wave velocity (aPWV) has been associated with mortality in clinical but not general populations. Aortic PWV was measured at baseline in 2488 older adults. Over an average of 4.6 years of follow-up, 265 deaths occurred with 111 categorized as cardiovascular in cause. Higher aortic aPWV quartile was associated with both total mortality (P = 0.010) and CV mortality (P=.006), independent of age, gender, race, SBP, known CV disease, creatinine, cholesterol and smoking. Thus, among a generally healthy, well-functioning, community dwelling population, aPWV, a marker of arterial stiffness, is associated with higher total and CV mortality (Circulation 2005;111(25):3384-3390). 2. The prevalence of the metabolic syndrome, a potent risk factor for cardiovascular diseases, has not been adequately explored in older individuals. Moreover, 2 sets of criteria have been proposed for the definition of the metabolic syndrome, one by the World Health Organization (WHO) and one by the National Cholesterol Education Program (ATPIII). We found that the prevalence of this syndrome in a subgroup of 2175 older participants from the Cardiovascular Health Study (CHS) free of cardiovascular disease at baseline to be 28.1% by ATPIII criteria, and 21.0%, by WHO criteria. The two sets of criteria provided concordant classification for 80.6% of participants. Multivariate Cox propotional hazard models showed that the metabolic syndrome defined with the ATPIII criteria, but not with the WHO criteria, was an independent predictor of coronary or cerebrovascular events, and was associated with a 38% increased risk (HR 1.38, 95%CI 1.06-1.79, p< 0.01). Thus, as defined by the ATPIII criteria, the metabolic syndrome yields independent prognostic information, even after adjusting for traditional cardiovascular risk factors and the individual domains of the metabolic syndrome (Diabetes Care 2005;28(4):882-887). 3. Increased thickness and stiffness of large arteries may contributeto why aging is the most important risk for cardiovascular diseases. Several large studies have shown that moderate alcohol consumption reduces the risk for heart disease and stroke. We examined whether alcohol consumption alters age-associated increases in arterial stiffness and intimal medial thickness (IMT). A total of 563 volunteers from the Baltimore Longitudinal Study of Aging had carotid duplex ultrasonography with measurements of IMT and an arterial stiffness index. Alcohol intake was assessed by a standard questionnaire. A U-shaped relationship was found between alcohol intake and stiffness index, with the lowest index in moderate drinkers; this relationship persisted after adjustment for cardiovascular risk parameters. Moderate drinkers showed 50% less age-associated increase in arterial stiffness than heavy drinkers and nondrinkers, both before and after adjusting for other cardiovascular risk factors. A significant positive u-shaped relationship was found between alcohol intake and IMT, which did not persist after adjustment for risk factors. Thus, light to moderate alcohol intake favorably modulates aging of the arterial tree. This effect may explain in part the J- or U-shaped relationship between alcohol intake and cardiovascular disease (Am J Cardiol. 2005;95:1006-10). 4. Long-term beta blocker therapy improves outcomes in patients with congestive heart failure (CHF). Outpatient beta blocker titration usually requires frequent clinic visits to monitor hemodynamics and symptoms. We developed an automated voice-interactive telemedicine system to assess symptoms, pulse, and blood pressure using home monitoring devices with information relayed via the telephone to a computer database which could be accessed by health ccare providers. Forty-nine patients with CHF not receiving beta-blockers were randomized to a TeleWatch assessment guided (TG) or outpatient clinic assessment guided (CG) carvedilol titration schedule with a goal of 25 mg BID over a three month period. Titration decisions were made by physicians blinded as to whether the information provided to them was obtained using the TG or CG data. The titration time was significantly shorter in the TG, compared to the CG group, 32.4 days versus 67.8 days, p< 0.001. On multiple linear regression analysis, group assignment (p< 0.001), final carvedilol dose (p=0.02) and NYHA class (p=0.04), were independently correlated to the duration of titration, while age, ejection fraction, pulse and systolic blood pressure were not. Thus, an automated telemedicine system can safely and significantly decrease the time to achieve carvedilol titration in CHF outpatients (Am Heart J 2006;151:844.e1-10. 5. Numerous studies have implicated trait anger and hostility in the pathogenesis of coronary heart disease. We investigated the associations of anger frequency (trait anger), expression (anger-out) and suppression (anger-in) with carotid artery intimal medial thickness and stiffness in younger and older healthy volunteers from the Baltimore Longitudinal Study of Aging. We found that anger suppression is an independent determinant of carotid arterial stiffness, but we did not observe any association between anger and carotid arterial thickness. Whether behavioral interventions focusing on anger expression can be effective in preventing or reversing arterial stiffness and its attendant cardiovascular risks deserve further evaluation. (Am J Hypertens 2006;19:1129-34). 6. Arterial thickness and stiffness increase with advancing age, and are increasingly recognized as risk factors for cardiovascular morbidity and mortality. Because these complex traits are likely to be affected by a multiplicity of genetic and environmental factors, the search for novel genes that may underlie these phenotypes will require genome wide linkage and association studies. We evaluated the heritability of blood pressure and central arterial structure and function in a founder population, i.e. one with high degrees of interrelatedness among its individuals due to limited external admixture. We recruited 6,148 subjects (57% women, 711 families), representing over 60% of the total population aged 14-102, from a cluster of 4 towns on the island of Sardinia. Heritabilities were assessed with simple variance components models, which assume that all similarities between genetic factors are due to additive genetic effects. The narrow heritability estimates for systolic BP, diastolic BP, diastolic carotid diameter, intimal medial thickness, and pulse wave velocity (an index of central arterial stiffness), adjusted for sex, age and age2, are respectively 0.258, 0.187, 0.443, 0.185 and 0.226, and indicate that 18% to 44% of the variance can be attributed to genetic factors. Thus, in this study of a large founder population, indexes of arterial structure and function show robust heritability estimates, indicating that genome wide linkage and association studies will likely succeed in identifying genes that contribute to the variance in these traits (PLoS Genet. 2006;2(8):e32).